The project objective is to develop a bioinformatics foundation for deciphering the metabolic network of every organism with a fully sequenced genome, in support of drug discovery, metabolic engineering, systems biology, and basic science. Our approach is based on a gold-standard metabolic database, MetaCyc, which is curated by Ph.D.- level biologists, from the experimental literature. A second objective is to further develop BioCyc, an evolving collection of Pathway/Genome Databases for 5,000-10,000 sequenced prokaryotic genomes. BioCyc will become the premier source of prokaryotic genome data because of its planned comprehensive coverage of prokaryotic genomes; its integration of multiple information sources; its powerful and user-friendly bioinformatics search, visualization, and analysis tools; and its distribution of data via multiple access channels. We have four specific aims. (1) To expand MetaCyc, a highly curated multi-organism database of metabolic pathways and enzymes that serves as an encyclopedic reference of metabolic information. MetaCyc can be used to predict the metabolic pathway complement of an organism from its sequenced genome. Information about experimentally determined metabolic pathways and enzymes will be curated into MetaCyc from the biomedical literature, with a focus on prokaryotic, fungal, and plant information. (2) To computationally generate BioCyc, a collection of organism-specific Pathway/Genome Databases for completely sequenced prokaryotes and model organisms that includes predicted metabolic pathways, predicted metabolic pathway hole fillers, and predicted operons. (3) To enhance the Pathway Tools software that supports the querying, visualization, and analysis of MetaCyc and BioCyc to include new comparative genomics capabilities; to include genome-context-based predic- tions of functionally related proteins and of novel pathways; to include a tool for iteratively browsing the reaction neighborhood of a metabolite; and to provide textual searches against multiple BioCyc databases. (4) To make MetaCyc and BioCyc available to the scientific community through a Web portal and via downloadable data files and software.